Abstract
BACKGROUND: Medicaid enrollment has been associated with disparities in younger cancer patient survival. To further understand this association for central nervous system (CNS) tumor patients, we used Surveillance, Epidemiology, and End Results (SEER)-Medicaid-linked data to examine associations between Medicaid enrollment and enrollment timing and (1) diagnosis stage, and (2) CNS tumor death. METHODS: Individuals diagnosed with a first malignant primary CNS tumor between 0 and 39 years from 2006 to 2013 were included. Medicaid enrollment was first classified as enrolled versus not enrolled with those enrolled further classified as having continuous, discontinuous (at diagnosis or other discontinuous), or other enrollment. We used logistic and Cox Proportional Hazards regression stratified by age to calculate adjusted odds ratios (ORs) and hazard ratios (HRs) for those 0-14 and 15-39 years. RESULTS: Among 10 107 CNS tumor patients, we found significantly higher odds of regional/distant versus in situ/localized stage diagnoses for those with other discontinuous (OR(0-14) = 1.50, 95% CI: 1.15-1.95) and at diagnosis (OR(15-39) = 1.41, 95% CI: 1.11-1.78) Medicaid enrollment versus those not enrolled. Those enrolled versus not enrolled in Medicaid had a higher hazard of CNS tumor death for both age groups (HR(0-14) = 1.60 95% CI: 1.37-1.86; HR(15-39) = 1.50, 95% CI: 1.39-1.62) with the highest hazards for those enrolled at diagnosis (HR(0-14) = 1.83, 95% CI: 1.51-2.22; HR(15-39) = 1.93, 95% CI: 1.77-2.10). CONCLUSIONS: Medicaid enrollment is associated with a higher risk of CNS tumor death with an almost 2-fold higher risk for young CNS tumor patients enrolled at diagnosis. These results support the critical need for consistent health insurance coverage for young CNS tumor patients.